Pass rate is higher (back to normal range) because of changes in the arbitrary passing threshold, a less ridiculous exam, and anyone taking studying way wayyyyyy harder than people would have in previous years given the issue in 2018. I don't think it's a single factor, albeit the most important one was a decision to NOT have the pitchforks out again at ASTRO. I think that the 85% of questions requirement coming from primary sources helped. I will give some credit to the presence of a study guide, that while it did not have the information in detail, at least told you what topics you should really make sure you know from other sources (especially in Cancer Biology).
Also helped, IMO, that RadOncQuestions did a targeted revamp of their RadBio/Physics questions sections just in time for test takers this past year.... there were upwards of 1,000 questions on each of those topics created directly from Hall and McDermott, written by people proficient in English (which is more than what can be said for a small fraction of the physics/rad bio questions I remember).
That being said, All of Ken Oliviers points would be welcome additions, IMO. A synopsis for those who can't access his most recent tweets:
First time pass rate 98% for Rad Bio/Physics. For clinicals, 99%. Repeating examinees with a 93% pass rate (unclear if Rad Bio/Physics or across all 3).
Changes being proposed to testing schedules:
1) Allow PDs to approve residents to take rad bio/physics as early as PGY-2, meaning a failure may not delay things in the long run (good idea IMO)
2) Moving clinical exam to May/June of PGY-5 year rather than post residency as people initiate new jobs (another good idea IMO)